explaining japan’s revolving door premiership: applying

Politics and Governance, 2016, Volume 4, Issue 2, Pages 36-53 36

Politics and Governance (ISSN: 2183-2463) 2016, Volume 4, Issue 2, Pages 36-53

Doi: 10.17645/pag.v4i2.575

Article

Explaining Japan’s Revolving Door Premiership: Applying the Leadership Capital Index

Tina Burrett

Faculty of Liberal Arts, Sophia University, 102-8554 Tokyo, Japan; E-Mail: [email protected]

Submitted: 30 January 2016 | Accepted: 21 March 2016 | Published: 23 June 2016

Abstract The tenure of Japanese prime ministers is famously short. Between 2006 and 2012 Japan changed prime minister once a year. What factors can explain Japan’s revolving-door premiership? To explore this puzzle, this article applies the Leadership Capital Index (LCI) developed by Bennister, ’t Hart and Worthy (2015) to case studies of the nine Japanese prime ministers holding office between 2000 and 2015. Leadership capital is the aggregate of leaders’ political re-sources: skills, relations and reputation. The LCI thus allows analysis of the interplay between individual capacities and contextual conditions in determining leaders’ ability to gain, maintain and deploy power. The LCI is applied to answer two questions. Firstly, what accounts for the short tenure of many Japanese premiers? In which of the LCI’s three lead-ership dimensions do Japanese leaders lack capital? Secondly, what forms of capital allow some prime ministers to re-tain office for longer than average (>2 years)? In particular, the article analyses the leadership of Junichiro Koizumi (2001–2006) Japan’s longest serving prime minister since the 1970s, and incumbent Prime Minister Shinzo Abe, who has held office for three years since December 2012. As well as utilising the LCI to comparatively analyse the tenure of Japan’s prime ministers, this article tests the applicability of the Index beyond Western parliamentary democracies.

Keywords authority; Japan; leadership; prime minister

Issue This article is part of the issue “New Approaches to Political Leadership”, edited by Mark Bennister (Canterbury Christ Church University, UK).

© 2016 by the author; licensee Cogitatio (Lisbon, Portugal). This article is licensed under a Creative Commons Attribu-tion 4.0 International License (CC BY).

1. Introduction

The turnover in holders of Japan’s highest office is fa-mously rapid. From 2006 to 2012, six prime ministers served Japan in as many years. Japan’s revolving door premiership has been attributed to the weak capacity of the prime minister’s office. But what then accounts for the relatively long tenure and predominant leader-ship of premiers such as Junichiro Koizumi (2001–2006) and Shinzo Abe (2012–)?

To explore the varied fortunes of Japan’s leaders, this article applies the Leadership Capital Index (LCI) developed by Bennister et al. (2015) to case studies of the nine Japanese prime ministers holding office be-

tween 2000 and 2015.1 Leadership capital is defined as aggregate authority, composed of three dimensions: a leader’s skills, relations and reputation. The LCI pro-vides a tool for systematically comparing leaders across these three broad areas that combines quantitative and qualitative elements. It thus allows more nuanced analysis than models relying on a single methodological approach by capturing the interplay between individual capacities and contextual conditions in determining leaders’ ability to act (Bennister et al., 2015, p. 417).

Bennister et al. employ an analogy between capital and authority to illustrate the difference between mere office holding and exercising political leadership. Politi-

1 Shinzo Abe’s two premierships (2006–2007 and 2012–) are accounted separately throughout the article.


cal authority is analogous to financial capital in that some leaders have it, while others do not. Like capital, authority can be saved, spent purposefully or frittered away. Office holding concerns gathering and conserv-ing leadership capital, while leading requires spending it wisely and replenishing one’s stock. Exercising lead-ership entails keeping stakeholders invested while tackling difficult and complex problems (Burns, 2003, p. 20; Heifetz, Grashow, & Linsky, 2009, p. 4).

This article uses the LCI to systematically compare the authority of Japan’s prime ministers in order to an-swer two questions. Firstly, does a broader lack of lead-ership capital, rather than mere institutional weakness, account for the short tenure of many Japanese prem-iers? If so, in which of the three leadership dimensions composing the LCI do Japanese leaders have low stocks? Do different Japanese prime ministers suffer from the same type of leadership deficit? Secondly, what forms of capital allow some Japanese prime ministers to retain office for longer than average (>2 years)? What factors allowed Junichiro Koizumi to become Japan’s second longest serving prime minister under the 1947 Constitu-tion? Why is incumbent Prime Minister Shinzo Abe on his way to beating Koizumi’s record, despite the fact that his first premiership ended ignobly after just one year? As well as utilising the LCI to comparatively analyse the authority of Japan’s prime ministers, this article tests the applicability of the Index beyond Western parliamentary democracies. I conclude that with minor adaptations to account for bureaucratic, factional and one-party domi-nant nature of Japanese politics, the LCI is a useful tool for robust analysis of political leadership in Japan.

In the sections that follow, I first examine existing scholarship on prime ministerial leadership in Japan. I then discuss the theoretical underpinnings of the con-cept of ‘leadership capital’, before presenting its three main forms (skills, relations and reputation capital). Next I introduce the Leadership Capital Index (LCI) and apply it to my nine prime ministerial subjects. I con-clude by summarising my answers to the two research questions outlined above and discussing further devel-opment of the LCI.

2. Leadership and the Authority of the Japanese Prime Minister

Under the 1947 constitution, 32 prime ministers have held office in Japan, among whom only five have spent four or more years in office. Owing to their short tenure, for most of the post-war period, Japanese prime minis-ters were considered institutionally weak (Shinoda, 2011, p. 48). Until the 2000s, the policy-making role of the Japanese prime minister was largely overlooked, with attention instead focusing on Japan’s powerful bu-reaucracy and factional politics within the ruling Liberal Democratic Party (LDP) (Krauss & Nyblade, 2005, p. 357). Aurelia George Mulgan used the term ‘un-

Westminster’ system to describe Japan’s policy-process:

“The [Japanese] system does not produce strong cabinet government with a prominent leadership role played by the prime minister, but a dual power structure of party-bureaucracy policy-making in which the prime minister and cabinet play a subordi-nate, rather than a superordinate role.” (2003, p. 84)

The unusually long premiership of charismatic Junichiro Koizumi from April 2001 to September 2006, however, led to a reassessment of prime ministerial authority (Machidori, 2006). Yet after raising expectations that his five immediate successors could not fulfil, Koizumi’s strong leadership was recognised as an exception ra-ther than the new norm (Mishima, 2012; Shimizu, 2005; Uchiyama, 2010). Shinzo Abe’s dominance since returning to the premiership for a second non-consecutive term in December 2012, however, sug-gests that Koizumi’s strong leadership was not unique and that the leadership capacity of Japan’s highest of-fice should be reconsidered (Burrett, forthcoming).

An increased potential for prime ministerial leader-ship within the Japanese government from the 1990s is attributed to changes in political communications, electoral reforms and/or a greater concentration of re-sources in the prime minister’s office (Kabashima & Steel, 2012; Krauss, 2000; Krauss & Nyblade, 2005; Otake, 2006). Changes in the relationship between pol-itics and the media in Japan have led to a greater focus on the prime minister (Krauss, 2000). Japanese prime ministers such as Junichiro Koizumi and Shinzo Abe successfully exploited new media dynamics to advance their personalised political agendas (Burrett, forthcom-ing). Changes in political communications allow leaders to circumvent their parties and appeal directly to vot-ers, a process enhanced by electoral reforms intro-duced in 1994.

Under Japan’s multi-member-district postwar elec-toral system, large parties were forced to run more than one candidate per constituency, producing incen-tives for intraparty factionalism. Each voter would cast just one ballot, but in a district that would send on av-erage three-to-five members to the House of Repre-sentatives, the lower house of Japan’s bicameral par-liament. Any party seeking to win two or more seats in a district would have to nominate multiple candidates. Intraparty factions helped to oil the wheels of a system of parallel party machines by providing patronage and financial support to different candidates within the same district (McCall, Rosenbluth, & Thies, 2010, 2010, pp. 55-56). The institutionalizing of factions within the LDP, that ruled Japan from 1955 to 1993, constrained the policy- and appointment-making powers of the party leader. To win the LDP presidency, and thus nom-ination to the post of prime minister, required a large factional power base. Changes introduced to the Japa-


nese electoral system in the wake of the LDP’s defeat in 1993 general election were designed to weaken fac-tions’ influence over both district election campaigns and national leader selection. The lower house elec-toral system was altered from a multi-member-district, Single Non-Transferable Vote (SNTV) system, to a com-bination of 300 single-member-districts and 11 propor-tional representation (PR) constituencies (180 seats). Under the new electoral system—used for the first time in 1996—political competition takes place along party lines (Krauss & Pekkanen, 2004). As a conse-quence, factions have lost control of the LDP leader-ship selection process, giving the prime minister poten-tially greater freedom over policy-making and cabinet appointments (McCall et al., 2010, p. 110).

Changes to the Japanese electoral system not only altered how the LDP selects its leaders, but also reor-dered the wider party system (Schoppa, 2011). Un-doubtedly, developments in the party system have had important consequences for the power and agency of the Japanese prime minister, not least in eventually bringing to office three premiers from the Democratic Party of Japan (DPJ). The PR element of the new elec-toral system has further encouraged fragmentation in a party system already prone to splits, mergers and de-fections. Colleagues with personal, policy or political grudges against the prime minister are easily tempted to jump ship, knowing that the PR ballot will act as a lifesaver to any new party that can attract a few per cent at the polls. The prevalence of personality politics in Japan—initially encouraged by the multi-member-district system—further exacerbates fragmentation, and along with the PR ballot, serves to entice populist mavericks to establish their own political parties. Splits and defections ordinarily reduce confidence in the prime minister, undermining his leverage with what remains of his party. Electoral reforms have thus been a double-edged sword for the authority of the Japa-nese prime minister over his own party.

The potential authority of Japan’s prime minister relative to that of other political actors has also been strengthened by a series of administrative innovations from the late 1990s that have shifted the locus of poli-cy-making from Japan’s powerful ministries to the cab-inet office. To reinforce the authority of the prime min-ister and the cabinet, in 1996, Prime Minister Ryutaro Hashimoto formed the Administrative Reform Council. The reforms subsequently passed by the Diet in July 1999 enacted significant institutional changes to strengthen the power and function of the cabinet sec-retariat—the prime minister’s support staff, equivalent to the British prime minister’s office (Shinoda, 2005, pp. 800-801). During his premiership from 2001 to 2006, Junichiro Koizumi introduced further reforms aimed at strengthening the policy-making power of the prime minister vis-à-vis Japan’s bureaucracy (Mishima, 2007, p. 727). From 2009 to 2012, the DPJ government

added its own reforms of bureaucratic power. Margari-ta Estévez-Abe argues that institutional changes in Ja-pan over the past two decades have cast the die in fa-vour of a Westminster system that centralizes power in the hands of the party leadership and prime minister, leading to ‘the Britannicization of Japan’ (2006, p. 633).

Yet in Britain, between 2000 and 2015, only three prime ministers held office, compared to three times that number in Japan. Despite developments enhanc-ing the potential authority of the Japanese prime min-ister since the 1990s, few incumbents manage to main-tain power for long. What light can the concept of ‘leadership capital’, explained below, shine on the weak authority and subsequent short tenure of many Japanese prime ministers?

3. Leadership Capital

In their original article, Bennister et al. draw on Pierre Bourdieu’s conceptualising of varieties of capital (eco-nomic, cultural, social, political) in constructing their theory of leadership capital (2015, pp. 418-419). For Bourdieu, political capital is largely symbolic. It is the process by which ‘agents confer on a person…the very powers they recognise in him’, giving that person credit to ‘impose beliefs’ and ‘recognised principles’ (in Ben-nister et al., 2015, p. 418). John Thompson similarly conceptualises political capital as a form of symbolic power that imbues its holder with ‘the capacity to in-tervene in events, to influence the actions and beliefs of others and indeed to create events, by means of the production and transmission of symbolic forms’ (2000, p. 98). In exercising symbolic power, leaders draw on various kinds of resources, including their reputation, popularity and accumulated prestige, assets that Thompson terms their symbolic capital (Thompson, 2000). Thompson argues that the use of symbolic pow-er is not incidental or secondary to the struggle for po-litical power, but essential to it:

“Anyone who wishes to acquire political power or to exercise it in a durable and effective fashion must also use symbolic power to secure the support of others within the political sub-field and within the broader political field.” (2000, p. 102)

Bourdieu and Thompson’s theorising enables the iden-tification of three elements of political power that are central to Bennister et al.’s concept of leadership capi-tal and to the categories composing the LCI that they develop to operationalise this concept (Figure 1).

Firstly, leaders must have the personal political skills to gain and maintain political power (Bourdieu, 2005, p. 39; King, 1991, p. 42). Leaders’ personalities and styles matter to the outcome of the policy process. Harold Lasswell (1936) pioneered the use of leadership typologies to understand the relationship between


personality and leaders’ performance. James Barber (1972) and Richard Neustadt (1991) offer typologies of presidential leadership in the United States. Also focus-ing on the U.S., Fred Greenstein (2001) provides a six-point framework for evaluating leaders’ political and personal skills based on: (1) proficiency as a communica-tor (2) organisational capacity (3) political skills (4) policy vision (5) cognitive style and (6) emotional intelligence.

Figure 1. Components of Leadership Capital. Source: Bennister et al. (2015, p. 422).

Second, political power is relational. Bourdieu saw po-litical power as derived from public trust granting a leader the capacity to mobilise supporters (In Bennister et al., 2015, p. 419). Leadership is an interaction be-tween individuals, rather than action by one individual. As Robert Tucker writes, ‘Leadership is a process of human interaction in which some individuals exert or attempt to exert, a determining influence on others’ (1981, p. 11). James MacGregor Burns’ definition of leadership similarly focuses on the interaction between leaders and their followers:

“Leadership over human beings is exercised when persons with certain motives and purposes mobi-lise, in competition or conflict with others, institu-tional, political, psychological, and other resources so as to arouse, engage and satisfy the motives of followers.” (1978, p. 18)

Relational Capital (R1) thus refers to the loyalties leaders mobilise, not only among voters but also among party colleagues, the media, business elites, bureaucrats and others possessing their own forms of capital that shape the leadership environment (Figure 1). Why people fol-low, or at a minimum accept a leader, shapes the nature of leadership authority. Burns argues that interaction between leaders and their followers takes two funda-mentally different forms. The first form Burns calls transactional leadership (1978, p. 19). The transactional leader interacts with followers for the purpose of ex-change. The exchange can be economic or political: the trading of votes between candidate and citizen, or the swapping of goods. The purposes of each party are re-

lated, at least to the extent that each party stands to gain from the exchange. Beyond this exchange, howev-er, the parties have no wider purpose holding them to-gether. Transactional leaders accumulate capital through technical competence and ‘bringing home the bacon’, rather than through a mobilising narrative. Burns calls the second form of interaction between leaders and fol-lowers transforming leadership (1978, p. 20). The trans-forming leader interacts with followers in a way that changes them both. For Burns, transforming leadership is inspirational leadership that gathers capital by mobilis-ing followers behind a particular vision or set of ideals.

Essential to leader-follower relations are followers’ perceptions of a leader’s skills. In the LCI, leaders’ Skills Capital is separated into ‘soft’ (S1) and ‘hard’ (S2) skills. Hard skills are instrumental and transactional, while soft skills concern inspiration, persuasion and shaping the preferences of others (Nye, 2008). Perceptions of a leader’s skills also relate to his or her Reputational Capi-tal (R2). A leader’s reputation is determined by their own behaviour and its observable impact, but also through interpretation of their behaviour by followers, colleagues, critics and other observers (Greenstein, 2000, p. 182). By spending political capital, leaders can reinforce, alter or destroy their reputation, with im-portant consequences for their future authority (Dahl, 1961, p. 229). Leadership capital increases when a lead-er’s reputation meets two conditions: when its norma-tive nucleus is considered appropriate for the times; and when the gap between political promises and perfor-mance is slight (Bennister et al., 2015, p. 423; Skow-ronek, 1993). Leaders’ reputations are most effective when their personal biography, political philosophy and in-office decisions are widely perceived to align.

Analysing political leadership through the prism of leadership capital assumes that it is the interaction be-tween an individual’s personal capabilities and their in-stitutional and situational environment that deter-mines a leader’s ability to act. Prime ministers, however powerful, are dependent on institutional structures and on the wider context in which they op-erate (Heffernan, 2003, p. 368). Although all leaders must work within constraints, some possess personali-ties more disposed to challenging these constraints than others (Herrmann, 1988; Keller, 2005). The man-ner in which different prime ministers approach similar situations varies significantly depending on their indi-vidual character, as Anthony King observes:

“Different people bring different personalities to the job; they have different goals; they adopt different styles; and they find themselves operating in different political environments. Second, there is variety within the lifetime of a single premiership.” (1991, p. 42)

There can be no doubt that Junichiro Koizumi’s crusad-er style of leadership varied wildly from Yasuo Fukuda’s


more managerial approach. Nor can it be denied that Shinzo Abe’s governing style was more dominant at the start of his second premiership than during his first. What is more difficult to discern is the extent to which policy and other political outcomes are determined by individual capabilities (oratory, charisma, negotiation) versus context. Prime Minister Yoshihiko Noda had lit-tle choice but to take a conciliatory approach to work-ing with Japan’s powerful bureaucracy and the divided factions within his own party, given the backdrop to his premiership—a lack of personal mandate, the ongoing nuclear crisis at the Fukushima nuclear plant and eco-nomic stagnation. But Noda’s conciliatory approach may also have been the case in any context. In his speech to the party caucus that made him prime minister, Noda described himself as more of a loach—a type of bottom feeding fish—than a ‘goldfish in a scarlet robe’ (Hayashi, 2011). Noda’s self-description suggests he was a concili-ator by nature and not just owing to circumstance.

Robert Elgie (1995) offers an interactionist model for the study of political leadership, combining person-al and systemic elements. His approach, designed for comparative analysis of leadership across liberal democ-racies, supposes that ‘political leaders operate within an environment which will both structure their behaviour and constrain their freedom of action’ (1995, p. 8). Elgie supposes that leaders possess agency—the capacity to act independently—allowing them to shape the envi-ronment in which they operate, if only up to a point, to improve their chances of success. The LCI similarly un-derstands political authority as the product of a leader’s perceived skills and the environment in which they op-erate. Capital accumulated or spent in one area has an impact on other elements of leadership capital.

4. The Leadership Capital Index (LCI)

The LCI is a diagnostic checklist for analysing a leader’s stock of authority, designed to identify variations in the aggregate level of leadership capital (Bennister et al., 2015, p. 423). In this study, the LCI is applied compara-tively to nine Japanese prime ministers to create a leadership league table (Table 3). It is used to provide a snapshot of the political authority of each leader at the mid-point of their premiership. Ideally, snapshots would be taken at various intervals during each leader’s tenure, as authority tends to ebb and flow over time (Breslauer, 2002, p. 13; Bynander & Hart, 2006). Looking at only one moment in a leader’s tenure could potentially create a biased picture. But for the nine leaders analysed here, selection bias is limited by the short tenure of most sub-jects. Seven of the nine leaders were prime minister for less than 15-months. To mitigate bias, for the longest serving leader, Junichiro Koizumi, three snapshots are analysed, at the start, middle and end of his five-and-a-half-year premiership (Table 5).

The LCI assesses leadership capital as an aggregate

of skills, relations and reputation and has the potential to provide a more nuanced picture of a leader’s au-thority than approval ratings or other quantitative data alone (Table 1). The variables included relate to the three elements of leadership capital defined above. Many indicators relate to perceptions and are meas-ured using a mixture of ‘hard’ empirical data (public opinion polling, election results) and ‘soft’ interpretive assessments (expert panels, political biographies, me-dia reports). This follows the approach used by Bennis-ter et al. in their original article and is in line with an emerging mixed methods paradigm (Bennister et al., 2015, p. 245; Burke Johnson & Onwuegbuzie, 2007; Hesse-Biber & Burke Johnson, 2013, 2015). The sources of LCI measurement are outlined in Table 2.

I have operationalised the Index using nine of the ten criteria suggested by Bennister et al. (2015, p. 424). The indicators included in the original LCI, ‘chosen by a process of reduction, distilling a vast array of variables often used to access political leadership down to a manageable number of 10’, are appropriate for as-sessing leadership in any parliamentary democracy, in-cluding Japan (Bennister et al., 2015, p.425). To better fit the context of Japanese politics, however, I have re-placed one indicator—that measuring public trust in a leader—with a measure of perceived relations between the leader and the Japanese bureaucracy. I have made this change for two reasons. First, in Japan, public opin-ion data on trust in politicians is not collected separately from personal approval ratings. As personal ratings poll-ing data is used to measure public perceptions of a lead-er’s skill (S2), it would not be appropriate to use this da-ta a second time to populate another indicator.

Second, relations with the bureaucracy are an im-portant factor determining Japanese prime ministers’ ability to exercise authority. In most parliamentary democracies, governments operate under the conven-tion of ministerial responsibility. Ministers take respon-sibility for the activities of their departments based on the notion that authority rests with elected politicians. Public servants follow ministers’ instruction and are ac-countable to them. Ministers in turn answer to parlia-ment and the public for everything that happens within their departments. In Japan, bureaucrats do not regard themselves as accountable to their ministers and there are few mechanisms through which to enforce ac-countability. Rather, Japanese bureaucrats largely con-sider themselves an independent source of political au-thority (George Mulgan, 2000, p. 187). Bureaucrats often have their own agendas and do not, as a matter of course, follow the instructions of ministers. Unlike in other parliamentary democracies, where publically de-fending policy is the duty of the minister, Japanese bu-reaucrats openly advocate particular policy positions, sometimes those contrary to their minister’s position. Bureaucrats can even answer questions on the floor of the Japanese parliament in place of ministers.


Table 1. Leadership capital index of a Japanese prime minister.

Criteria Indicator Measurement (score of 1 low to 5 high)

S1 01 Political/policy vision (1) Completely absent (2) Unclear/inconsistent (3) Moderately clear/consistent (4) Clear/consistent (5) Very clear

S1 02 Communication skills (1) Very poor (2) Poor (3) Average (4) Good (5) Very good

S2 03 Personal poll rating (1) Very low (< 20%) (2) Low (20-34%) (3) Moderate (35-49%) (4) High (50-64%) (5) Very high (> 65%)

S2 04 Longevity (time in office) (1) <1 year (2) 1-2 years (3) 2-3 years (4) 3-4 years (5) >4 years

S2 05 (Re)election as party leader (margin) (1) Very small (<1% of electors) (1) (2) Small (1-5%) (3) Moderate (5-10%) (4) Large (10-15%) (5) Very large (>15%)

R1 06 Party polling relative to most recent election result

(1) < -10% (1) (2) -10 to -2.5% (2) (3) -2.5% to 2.5% (3) (4) 2.5 to 10% (4) (5) >10% (5)

R1 07 Likely serious challenge at next party presidential election

(1) Very high (2) High (3) Moderate (4) Low (5) Very low

R1 08 Working relations with the bureaucracy (1) Very poor (2) Poor (3) Average (4) Good (5) Very good

R2 09 Perceived ability to shape party policy platform

(1) Very low (2) Low (3) Moderate (4) High (5) Very high

R2 10 Perceived parliamentary effectiveness (1) Very low (2) Low (3) Moderate (4) High (5) Very high


Table 2. Source of LCI measurement.

Criteria Indicator Measure

S1 01 Political/policy vision Soft (expert)

S1 02 Communication skills Soft (expert) S2 03 Personal poll rating Hard (polling)

S2 04 Longevity (time in office) Hard (chronology)

S2 05 (Re)election as party leader (margin) Hard (vote count)

R1 06 Party polling relative to most recent election result Hard (polling)

R1 07 Likely serious challenge at next presidential election Soft (expert)

R1 08 Working relations with the bureaucracy Soft (expert) R2 08 Perceived ability to shape party policy platform Soft (expert)

R2 09 Perceived parliamentary effectiveness Soft (expert)

Table 3. LCI measure of Japanese prime ministers 2000–2015.

Mori Koizumi Abe 1 Fukuda Aso Hatoyama Kan Noda Abe 2i

S1 01 Political/policy vision 1 4 2 2 1 2 2 2 4 S1 02 Communication skills 1 5 2 2 1 1 2 3 2 S2 03 Personal poll ratingii 2 4 3 3 2 4 2 2 4 S2 04 Longevity (time in office) 2 5 1 2 1 1 2 2 4 S2 05 (Re)election as party leader (margin)

NA (2)iii 5iv 5 5 5 4 5 4 4v

R1 06 Party polling relative to most recent election resultvi

2 2 3 3 3 2 2 1 3

R1 07 Likely serious challenge at next presidential election

1 5 2 2 1 2 1 2 5

R1 08 Working relations with the bureaucracy

3 4 3 4 3 2 2 3 4

R2 09 Perceived ability to shape party policy platform

1 4 2 2 2 1 2 2 4

R2 10 Perceived parliamentary effectiveness

1 4 2 2 1 1 1 2 4

Total score 16 42 25 27 20 20 21 22 38

Notes: i Abe was the incumbent prime minister at the time of writing. Analysis of his second premiership is up to and in-cluding December 2015; ii Monthly personal poll ratings are given as average for each premiership (see Appendix); iii Mori was elected party leader unopposed in a emergency ballot after Prime Minister Obuchi died following a stroke. Mori’s to-tal score has been adjusted based on average score for the other nine indicators; iv Koizumi first won election as LDP presi-dent in April 2001. He was re-elected unopposed in August 2001. He won a third clear victory against three challengers in September 2003; v In 2012, Abe lost the first round presidential ballot in which both party chapters and parliamentary rep-resentatives could vote to Shigeru Ishiba. He won the second round run-off ballot in which only parliamentarians can vote by 54.8 per cent to Ishiba’s 45.2 per cent. Abe was re-elected as party president unopposed in September 2015. The score given here averages his performance over these two presidential elections (Abe would receive a score of 3 in 2012 and 5 in 2015); vi In Japan both the upper and lower houses of parliament are elected in national elections. Upper house elections take place every three years, with half the seats up for election each time. Lower house elections take place at least every four years, with the prime minister having the power to dissolve parliament. Japanese elections use a mixed electoral system, with both single member districts (SMD) and a PR ballot. Election results used here combine both SMD and PR ballots.

In addition to including bureaucratic relations, I have made one other partial change to the LCI offered by Bennister et al.. In their original, to measure public perceptions of a leader’s skill (S2), the leader’s current personal poll ratings are presented relative to that leader’s ratings at the most recent election (Bennister et al., 2015, p. 424). In Japan, very few prime ministers come to office at a general election. In fact just five of the 32 prime ministers that have served under the

1947 post-war constitution came to office this way. Among the nine prime ministers studied here, just two—Yukio Hatoyama and Shinzo Abe—took office as the result of a general election (see Appendix).2 The other seven all came to power part way through par-

2 Junichiro Koizumi came to power in April 2001 after the res-ignation of Yoshiro Mori, but subsequently won the 2003 and 2005 general elections.


liament after their predecessor resigned. In the Japa-nese case, comparing a prime minister’s personal rat-ings to those at the last election is inappropriate, as in the majority of cases a different leader was in office at that time. To capture public perceptions of a leader’s skills, I use personal approval polling data without comparison to ratings at the last election (Table 2).

5. Applying the LCI: Japan’s Prime Ministers 2000–2015

The following analysis uses the LCI to answer the two research questions posed in the introduction to this ar-ticle. Analysis is based on hard data combined with in-sights from biography, media reports and my personal interviews with Tokyo-based politicians, political advi-sors and journalists.

5.1. Explaining Short Tenure

Among the nine prime ministers analysed here, seven served less than 15-months in office (Table 3). Applying the LCI reveals that the leadership of short-serving prime ministers has several features in common. All suffered from a lack of policy vision and an inability to shape their party’s platform that severely limited their authority. In each case, these leaders failed to offer ei-ther transformational or transactional leadership. They were neither able to gather capital through mobilising ideals and aspirations, nor by ‘delivering the goods’.

In most cases, weak communication skills exacer-bated leaders’ inability to offer a coherent personal and policy narrative. Although all seven prime minis-ters came to office with healthy personal approval rat-

ings, these quickly evaporated when they failed to ar-ticulate a clear purpose in seeking power. Each saw his personal approval ratings drop by 30 per cent or more between the start and end of his premiership (Figure 2).

Prime Minister Yoshiro Mori ended his premiership with approval ratings of single digits and is the lowest scoring leader on the LCI (Table 3). Among the leaders studied here, Mori is a special case, coming to office af-ter sitting Prime Minister Keizo Obuchi suffered a stroke. As LDP general secretary, Mori was elected prime minister unopposed in an emergency vote. He is mainly remembered for his gaffes, scandals and undip-lomatic comments. Even before becoming prime minis-ter, Mori was described in the Japanese media of hav-ing ‘the heart of a flea and the brain of a shark’ (BBC, 2000). Without Obuchi’s sudden death, Mori would not have become prime minister, which helps explain his exceptionally low LCI score.

LDP Prime Ministers Abe (in his first term), Fukuda and Aso were undone by divisions within their party as well as by their personal leadership deficiencies. Policy ruptures related to vested interests within the LDP hampered efforts to tackle Japan’s stagnant economy, the top priority of Japanese voters (Mishima, 2012, p. 278). All three leaders lacked the party management skills and direct popular support to overcome obstacles to economic reforms erected by members of their own party. Abe appointed his personal friends to key posi-tions within the cabinet and bureaucracy, but failed to control them when bitter infighting occurred. Fukuda’s technocratic nature often left him as a bystander in policy debates. Aso’s tenancy to make flippant remarks and privileged personal background made him look out of touch.

Figure 2. Prime minister’s personal approval rating by month in office (%). Source: NHK (2015). Retrieved from

http://www.nhk.or.jp/bunken/yoron/political/index.html


Ko Mishima (2012) notes structural reasons for the dearth of leadership ability at the top of the LDP in the 2000s, citing the disappearance of the traditional ca-reer path to the premiership. In its earlier days, promo-tion within the LDP was based on seniority, allowing politicians to systematically build the knowledge and skills necessary for policy-making (Mishima, 2012, p. 281). Politicians had to prove themselves able before reaching the cabinet. Furthermore, when intraparty fac-tions controlled leadership selection, only those with good brokering skills could win the LDP presidency. As factional power has declined within the party since the 1990s, those without leadership skills and training can rise to the top. Abe and Fukuda came to the premiership after relatively short careers compared to their prede-cessors. Aso had a longer career, but his experience was narrower than that of most past prime ministers. Abe, Fukuda and Aso are all descendants of previous prime ministers, providing them with the networks to win the LDP presidency despite their leadership deficiencies.

The decline of faction-based leader selection has al-so left prime ministers more vulnerable in the face of declining personal approval ratings. Loyalties within the LDP and DPJ are now more fluid, leading party rep-resentatives to abandon a prime minister with falling public support. In the past, a prime minister was secure in office even if he lost public confidence, as long as he maintained his factional coalition (Matsumoto, 2001). Today, parliamentarians are more concerned with the reputation of their party leader than in the past, as since electoral reforms in 1994 introduced PR ballots and SMDs, voters focus more on the image of national parties than on the personal traits of their local candi-dates (Krauss & Pekkanen, 2011). The personal popu-larity of the national leader plays an important role in creating the party’s public image.

Fixed term elections for the party presidency en-courage challenges to an unpopular prime minister among his intraparty rivals (Takayasu, 2010).3 Prime ministers often choose to jump before they are pushed, resigning rather than facing a leadership chal-lenge they may not survive. Assuming office outside of the election cycle deprived most of the prime ministers studied here of a popular mandate, further weakening their authority and position vis-à-vis intraparty rivals. An exception is Yukio Hatoyama, who led the DPJ to election victory in August 2009, becoming head of the first single-party non-LDP government since 1955.

Despite his electoral mandate, Hatoyama served just 256 days in office, the shortest tenure of any prime minister included in this study (see Appendix). Hato-yama propensity for making rash policy pronouncement

3 The LDP holds presidential elections every three years (every two years until 2002). The DPJ held presidential elections every two years until 2011, but extended the presidential term to three years in 2012.

on television, without the political skills to realise them, proved his undoing. In particular, his u-turn on a promise to relocate the U.S. Futenma marine military base out-side Okinawa cost him public, press and party support. Hatoyama played an important role in launching the DPJ in 1996, largely owing to a personal fortune inherited from his mother. It was his financial rather than leader-ship capital that propelled him to the party leadership. The DPJ made most of its political gains under Hatoya-ma’s predecessor Ichiro Ozawa, who was forced to re-sign the leadership following a financial scandal just three months prior to the 2009 general election. Hato-yama was elected leader with Ozawa’s backing.

Like their immediate LDP predecessors—Abe, Fuku-da and Aso—the DPJ’s three prime ministers—Hatoyama, Kan and Noda—largely failed due to their in-ability to offer a convincing plan for economic recovery. The DPJ initially talked of implementing a social demo-cratic style ‘Third Way’ (Daisan no Michi) between tradi-tional LDP state-guided capitalism and neoliberalism, but quickly fell back on the latter. The three DPJ prime min-isters also lost public confidence by failing to implement many of the new spending programmes listed in the par-ty’s 2009 manifesto due to a lack of funds. Hatoyama and Kan were hampered in their efforts to find addition-al money by non-cooperation from the bureaucracy. Even prior to taking office, Hatoyama caused friction with officials by naming curtailment of bureaucratic power as his top priority. Ministerial-led decision-making exacerbated existing policy and personality divisions among ministers and the wider parliamentary party.

Although the DPJ does not possess formal factions like the LDP, informal groups of parliamentarians gath-er around potential party leaders. But these groupings hold limited power over their members and are only one factor influencing the outcome of party leadership elections (Schmidt, 2011). Hatoyama, Kan and Noda were not backed by a stable alliance of groups. Like their LDP counterparts, their support among parlia-mentary colleagues was predicated on their public popularity. Kan, for example, was elected to replace Hatoyama in June 2010 following media reports that he was voters’ preferred choice (Mishima, 2012, p. 290). Kan’s stance as the anti-Ozawa candidate was another factor in his success. Admired and reviled in equal measure, Ichiro Ozawa, known as the ‘shadow shogun’ for his skill in backroom dealing, was a source of party division for all three DPJ prime ministers (George Mulgan, 2015). Ozawa’s leadership challenge to Kan three months after the latter took the helm cre-ated a fatal schism within the party (Rebuild Japan Ini-tiative Foundation, 2013).

Beating Ozawa initially gave Kan new buoyancy in the opinion polls (Figure 2). But Kan quickly squan-dered his new mandate as DPJ president by mishan-dling tensions with China over the Senkaku/Diaoyu Is-lands following a skirmish between a Japanese Coast


Guard vessel and a Chinese fishing trawler on 7 Sep-tember 2010. Beset by scandals and misjudgements, Kan seemed to be already on his way out of office when the Great Tohoku Earthquake and resulting Fu-kushima nuclear disaster struck on 11 March 2011. If managed effectively, crises can boost a leader’s reputa-tion. In times of crisis, people look to their leaders to ‘do something’. Successful crisis management can turn mere politicians into statesmen. But when a crisis is mismanaged, their visibility makes leaders the obvious scapegoat (Boin & ’t Hart, 2003, p. 544). To turn a crisis into an opportunity, leaders must shape the way a cri-sis is perceived and present convincing plans to man-age its ramifications (Foley, 2009, p. 502).

When confronted by crisis, Naoto Kan struggled to provide a reassuring response. His difficulties in man-aging the Fukushima disaster became bound up with pre-existing doubts over his capacity for leadership. Under the glare of media focus, Kan’s infamous temper and tendency to micromanage unnecessary details be-came a source of public criticism. His preference for re-lying on a small inner circle of personal advisors strained his working relations with the bureaucracy, delaying the emergency response on the ground (Na-tional Diet of Japan, 2012). Opinion polls in the months following the disaster showed two-thirds of voters were disappointed with the Kan’s handling of the Fukushima crisis (The Economist, 2011a). Thus when Ozawa again tried to unseat Kan by conspiring with the LDP to pass a non-confidence vote in the lower house, the prime min-ister had no option but to resign to forestall the vote.

Taking power following Kan’s resignation in August 2011, the DPJ’s final prime minister, Yoshihiko Noda at-tempted to overcome party divisions by using seniority rather than political affiliation to appoint his cabinet. But despite his considerable skill at building consensus, Noda was not able to hold his party together. The DPJ had become, in the words of Diet member Akihisa Na-gashima, ‘like Afghanistan’: an ungovernable collection of tribes revolving around loyalty to a few chieftains ra-ther than to a common ideology (The Economist, 2011b). Noda faced opposition from Ozawa and Hato-yama’s supporters on tax increases, social benefits re-form, and Trans-Pacific free trade. But he proved will-ing and able to build alliances within the bureaucracy and opposition parties to bypass his intraparty oppo-nents. In June 2012 Noda successfully brokered a tri-partite agreement with the opposition LDP and New Komeito to raise consumption tax from 2014.

Passage of the consumption tax bill, however, came at a high price for the prime minister. Interpreting Noda’s bill as an attempt to move the DPJ to the right, 50 left-leaning parliamentarians resigned the party whip (The Economist, 2012). Although in recognition of his tactical victory and tenacity, Noda’s approval rating initially rose after passage of the consumption tax bill, his popularity soon declined as his party began to im-

plode (Inoguchi, 2013, p. 188). Smelling blood, the LDP withdrew their cooperation with Noda, redoubling their efforts to force a general election (The Economist, 2012).

Unlike the other short-tenured prime ministers studied here, Noda was more the victim of circum-stance than of his own failings. By any measure, Noda came to power in difficult circumstances, inheriting es-calating territorial tensions with China, a faltering economy and a cumulative annual government debt that had reached 200 per cent of GDP. The March 2011 earthquake, tsunami, and nuclear disaster, compound-ed Japan’s economic quagmire. Public disappointment with his DPJ predecessors also weighed heavily on Noda. Throughout his premiership, the DPJ lacked a majority in the House of Councillors, which was lost at the July 2010 election. To pass legislation aimed at speeding recovery from the 2011 disasters and tackling the deficit, Noda required support from the opposition LDP and New Komeito. But with public support for the DPJ languished at around 20 per cent, he lacked the rela-tional capital to bring the opposition to the negotiating table. LDP leaders were reluctant to reach legislative deals with Noda that could forestall the calling of a gen-eral election, which the LDP was confident it would win. The DPJ’s unpopularity further led to a spate of defec-tions by its own parliamentarians. By mid-November 2012, Noda had lost his majority in the House of Repre-sentatives, forcing him to call a general election.

5.2. Explaining Long Tenure

Since 2000, only two Japanese prime ministers have served more than two years in office (Table 3). Junichi-ro Koizumi was prime minister for five-and-a-half years from April 2001, finally resigning in September 2006 owing to party imposed term-limits, despite retaining high public approval (Figure 3). At the time of writing, Shinzo Abe was entering the third year of his second non-consecutive term as prime minister. Re-elected to the premiership at the December 2012 general election, Abe was previously prime minister for one year follow-ing Koizumi’s resignation in September 2006. Articula-tion of a clear policy vision and a related personal narra-tive are at the heart of both leaders’ success (Table 3).

Junichiro Koizumi won the LDP presidency on his third attempt in April 2001, only after a huge popular vote from grassroots members forced the hands of party bosses (Stockwin, 2008, p. 105). Koizumi used his mas-tery of television and the popular press to appeal to his party’s rank and file above the heads of LDP elders who abhorred his radical neoliberal agenda. He went on to win three national elections by appealing directly to the electorate with the campaign slogan ‘Change the LDP, Change Japan’. Koizumi’s share of the vote actually strengthened over the course of his time in office (Table 4). His long hair and natural charisma were refreshing in a political system characterised by convention.


Figure 3. Prime minister Koizumi’s personal approval ratings April 2001–September 2006 (%). Source: NHK (2015). Re-

trieved from http://www.nhk.or.jp/bunken/yoron/political/index.html

Table 4. General election results 2003 and 2005.

2003 2005

SDM Seats PR Seats SDM Seats PR Seats

Liberal Democratic Party 168 69 219 77

Democratic Party of Japan 105 72 52 61

New Komeito 9 25 8 23

Communist Party 0 9 0 9

Social Democratic Party 1 5 1 6

Source: Election Resources (2015). Retrieved from http://www.electionresources.org/jp

By Japanese standards, Koizumi’s domestic agenda was radical. Cabinet posts were to be allocated on merit and no longer by faction. Spending on public works was to be slashed, and government borrowing capped. Banks would have to acknowledge the full extent of their bad loans and then sort them out to get the economy moving again. Above all, Koizumi planned to privatise the Japanese postal service, which lay at the heart of the parasitic relationship between Japan’s pol-iticians, bureaucrats and interest groups.

Shinzo Abe’s first premiership was marred by his indecision on economic reform and by poor party management that led to a series of scandals involving his senior ministers. Ultimately, Abe’s leadership was dealt a fatal blow by the LDP’s defeat in the House of Councillors election in July 2007. Six weeks later, Abe announced his resignation. But in September 2012, a combination of his own actions, the right circumstanc-es and a bit of luck allowed Abe to regain his party’s leadership (Burrett, forthcoming). Poor health was the official reason given for his 2007 resignation. In staging his comeback, Abe’s PR team used his recovery to build a narrative of personal drive and discipline in the face of adversity.4 To prove his newfound vitality, once re-elected LDP president, Abe hit the ground running with a clear set of policy aims. Abe’s first premiership had lacked policy focus, but Abe 2.0 made ‘Abenomics’—his plan to revive the economy through fiscal stimulus,

4 Author’s interview with Japan Times journalist, June 2014.

monetary easing and structural reforms—his pro-gramme showpiece. In focusing on the economy and promising neo-liberal structural reforms, Abe bor-rowed from the Koizumi playbook. Supported by a much-improved PR operation that includes Koizumi’s image guru Isao Iijima, Abe and his economic plan be-came ever-present on Japanese TV screens. His insist-ence that a reluctant Bank of Japan reverse deflation with significant quantitative easing, contrasted posi-tively with the incumbent DPJ government’s muddled economic strategy.

Abe’s bold and theatrical behaviour drew lessons from Koizumi’s leadership. In 2005, Koizumi dramati-cally withdrew the whip from 37 LDP parliamentarians opposed to his privatisation of Japan Post before call-ing a general election in which he ran his own hand-picked candidates against his former colleagues (Mi-shima, 2007, p. 734). Koizumi, whose popularity and political authority had been waning prior to the elec-tion, was rewarded with a landslide (Figure 3). But Koi-zumi’s victory was less the result of public enthusiasm for privatisation than of support for his strong and de-cisive leadership. Likewise, Abe’s comeback and second term popularity owed more to voters’ approval of his bold advocacy of Abenomics than to support for the specifics of his plan.

Concentrating his efforts on reviving the Japanese economy allowed Abe to win back a majority for his government in upper house elections in July 2013. For Abe, this was a personal victory, as it had been on his


watch in 2007 that the DPJ had replaced the LDP as the largest party in the House of Councillors. But despite impressive gains in the July 2013 elections, in the up-per house Abe still relies on coalition partners, New Komeito, for his majority. Koizumi also governed in co-alition with Komeito throughout his premiership.5 But in neither case was coalition a major constraint on the prime minister. Associated with the Buddhist sect Soka Gakkai, Komeito is a pacifist party relying predominant-ly on the religious faithful for its votes. Yet, despite its pacifist leanings, Komeito remained in government with Abe after his reinterpretation of Japan’s peace constitution to allow for collective self-defence (Bur-rett, forthcoming).6 Similarly, Komeito continued in co-alition with Koizumi despite his decision to introduce legislation allowing the Self-Defence Forces (Japan’s military) to be deployed to Iraq in 2003. In both cases, defence reforms were controversial with the Japanese public (Ishibashi, 2007; Yoshida, 2014). But Koizumi and Abe were able to maintain their coalitions owing to their broader electoral appeal. Throughout their prem-ierships, both were personally more popular than their party, minimising the chances of a serious leadership challenge or backbench rebellion by government MPs.

The examples of Abe and Koizumi show how if

5 Koizumi inherited a coalition government in April 2001. He failed to turn his 60 per cent approval rating into a majority for his party in the November 2003 general election and continued to rely on coalition partners. Koizumi finally won a majority for his party in the lower house in the September 2005 general election, but continued to rely on New Komeito for a majority in the upper house. 6 Article 9 of Japan’s constitution bars it from using force to re-solve conflicts except in the case of self-defence. In July 2014, the Abe cabinet reinterpreted the constitution to henceforth allow Japan to fight overseas to aide its allies.

spent wisely, leadership capital in one area provides dividends in others. Personal popularity allowed both leaders to challenge the policy status quo within the LDP. Their bold actions reinforced public perceptions of their leadership as decisive, giving them a stronger hand in dealing with intraparty dissent. Koizumi in par-ticular faced strong opposition from within LDP ranks as he attempted to dismantle patronage and pork bar-rel networks that had maintained the party’s power, but constrained its policies, for decades. Koizumi, how-ever, managed to turn this opposition to his advantage. His war against post office privatisation rebels sealed his reputation as the slayer of vested interests, allow-ing him to leave office at the height of his authority, contrary to the downward trajectory of most premier-ships (Laing & ’t Hart, 2011).

Although Koizumi left office on a high note, did he leave a lasting legacy? Did he exercise leadership or merely hold office? Few leaders achieve as much as they hope or promise. Although evaluations of Koizu-mi’s premiership are generally favourable, there are reasons to question the extent of his accomplishments and the endurance of his legacy (Anderson, 2004; Shimizu, 2005). This conclusion does not suggest that there were not significant achievements. Koizumi left office undefeated at the polls, with strong economic performance underpinning his general election victo-ries. He also energised Japan’s dealings with the world, albeit controversially in his support for the U.S.-led ‘War on Terror’. Domestically, Koizumi introduced sub-stantial reforms to the state apparatus, which ultimate-ly strengthened his authority over parliament and his own party (Table 5). But facing stiff opposition from within the LDP, Koizumi delayed in embarking on his personal reform agenda until late in his premiership, curtailing his domestic legacy.

Table 5. LCI measure of Junichiro Koizumi over time (2001–2006).

K1 May 2001

K2 Jan 2004

K3 Sept 2006

S1 01 Political/policy vision 4 4 5

S1 02 Communication skills 5 5 5

S2 03 Personal poll rating 5 4 4

S2 04 Longevity (time in office) 1 4 5

S2 05 (Re)election as party leader (margin) 5 5 5

R1 06 Party polling relative to most recent election result 2 2 2

R1 07 Likely serious challenge at next presidential election 5 5 5

R1 08 Working relations with the bureaucracy 4 4 4

R2 08 Perceived ability to shape party policy platform 3 2 4

R2 09 Perceived parliamentary effectiveness 3 3 4

Total score 37 38 43


Stubbornly resisted by his own party, Koizumi’s re-forms were all partial and much delayed. In the face of opposition, bold proposals were diluted. Rather than the savings and insurance functions of Japan Post being abolished, they would continue as separate organisa-tions, and privatisation would be delayed to 2017 (The Economist, 2006b). But Koizumi began the break up of Japan’s ‘iron triangle’—big business, the bureaucracy and the LDP—making government a bit more account-able and efficient. In expelling 37 parliamentary rebels who voted against privatisation, and putting up allies to run against them in a snap election in 2005, Koizumi destroyed the old LDP.

Koizumi’s achievements in foreign policy were also partial. Koizumi moved to ‘normalise’ Japan’s foreign and security policy. Challenging the limits placed on Ja-pan by its post-war pacifist constitution, he sent refuel-ling tankers to the Indian Ocean and peacekeeping troops to Iraq (Ishibashi, 2007). He worked with the U.S. to make Japan less dependent on America’s mili-tary umbrella and shoulder more of the burden of its own defence, but stopped short of amending the con-stitution to assert Japan’s right to participate in collec-tive security (Pekkanen & Krauss, 2005). Despite his foreign and security policy successes, Koizumi marred his international reputation by visiting Tokyo’s Yasuku-ni Shrine, which commemorates 14 war criminals along with millions of Japanese war dead (Hiwatari, 2005). His actions not only damaged relations with key trading partners China and South Korea, but also made territo-rial disputes harder to settle, and hardened opposition to Japan’s attempt to gain a permanent seat on the UN Security Council (The Economist, 2006a). In short, Koi-zumi’s Yasukuni visits undermined his ambition to make Japan more ‘normal’—i.e. ensuring a presence in international political affairs equal to its economic sta-tus. In December 2013, Abe’s visit as prime minister to the controversial shrine similarly poured cold water on already cool relations with neighbouring states occu-pied by Japan in World War Two.

External forces limit all prime ministers; in the case of Koizumi, a major constraint was hostility to his re-form agenda from members of his own party. Koizumi used his personal popularity to outmanoeuvre his LDP opponents and trump the faction system that had hitherto controlled Japan’s prime ministers. He intro-duced institutional reforms that changed the way au-thority is accrued to the prime minister’s office. Thanks to these reforms, political advancement came to de-pend more on loyalty to the prime minister than to fac-tion. But battling his party ran down the clock on Koi-zumi, constraining what he could achieve during his two-term-limit as LDP president.

Koizumi did more than merely hold office. He may not have achieved as much lasting change as trans-formative prime ministers like Shigeru Yoshida and Hayato Ikeda, but a number of his reforms live on.

The economic policy of Japan’s current prime minis-ter, Shinzo Abe, contains elements of Koizumi’s lais-sez-faire approach, including a commitment to fur-ther deregulation to promote growth. Abe has also carried forward Koizumi’s ambition to amend consti-tutional restrictions on collective security. Perhaps most significantly, Koizumi’s powerful exercise of leadership changed the image of the prime minister in Japan. After Koizumi’s success, Japanese voters began to demand that the prime minister lead policymaking with more force. Some commentators predicted a new era of strong prime ministers (Machidori, 2006; Takenaka, 2006). But Koizumi’s immediate successors failed to live up to the expectations raised by his leadership. Koizumi proved a hard act to follow. His strong personal leadership hollowed out his party, leaving it in a state of disarray.

Koizumi paid little attention to building structures within the LDP to allow his revolution to continue be-yond his tenure as party president. He failed to build new policy groupings around his neoliberal agenda to replace the traditional LDP zoku (policy tribes) that his leadership undermined. A natural loner, he neglected to groom potential protégés to carry forward his re-forms. Ultimately, Koizumi’s personal leadership traits—especially his willingness to take on vested in-terests and challenge the status quo—proved more adept at destroying the old than creating the new. He lacked the bargaining skills to build new structures when confronted by opposition from other authorita-tive actors.

In his second term, Shinzo Abe has clearly learned from Koizumi’s mistakes, in particular by paying closer attention to party management than his predecessor. Abe has successfully remade the LDP in his own image, by selecting candidates sharing his right-wing ideology to contest the 2012 general election. The hundred-plus freshmen representatives who rode to power on Abe’s coattails in December 2012 have largely remained loyal to the prime minister. Unity within his parliamentary party has allowed Abe to pass controversial security and secrecy legislation at breakneck speed. In achiev-ing constitutional and security reform, Abe has gone further, and faster, than Koizumi.

Abe has also paid closer attention to political ap-pointments than Koizumi, who by temperament tend-ed to act as a lone wolf. During his first term, Abe staffed the prime minister’s office and cabinet with parliamentarians to whom he was personally close, giv-ing them roles such as ‘Special Advisor to the Prime Minister’. But Abe’s friends proved poor lieutenants. Decision-making became bottled-necked, and bitter turf wars ensued, as Abe’s inner circle jealously guard-ed their access to the prime minister (Burrett, forth-coming). A dysfunctional chain of command contribut-ed to Abe’s downfall. In his second term, Abe clearly demonstrated greater political skill in making appoint-


ments than during his first premiership. For his second government, Abe was quick to choose Yoshihide Suga as his chief cabinet secretary. His appointment reflects Abe’s recognition of Suga’s political ability, rather than a personal connection between the two men. Abe made it clear that Suga was the gatekeeper to the Prime Minister’s Office. This made Abe’s second gov-ernment much more effective in managing policy, par-liament, and public relations, than his first administra-tion. In comparison, Koizumi’s tendency for self-reliance limited what he was able to achieve.

Despite running a tighter ship second time around, Abe’s policy legacy to date is a mixed picture. Like Koi-zumi before him, Abe has delayed in introducing eco-nomic reforms opposed by vested interests within the LDP. Abe came to office promising massive structural reforms to boost economic growth. To his credit, Abe has tackled some of Japan’s most entrenched interest groups. Abe has broken the power of the agricultural cooperatives, bypassing their objections to sign the Trans-Pacific Partnership (TPP) trade deal. The prime minister has also reformed corporate governance with unexpected speed and determination. But labour re-form, desperately needed to raise productivity, re-mains untouched (Harding & Lewis, 2015).

Despite the fanfare, Abenomics has not revived Ja-pan’s economy. The BoJ’s monetary expansion and as-set purchases have pushed the value of the yen down to multi-year lows against the dollar and sparked a stock market rally that ran from just before Abe took office in late 2012 through most of 2015. But, the Jap-anese economy has performed unevenly, falling into recession in mid-2014 and then swinging between quarters of growth and contraction in 2015. Rather than focusing on badly needed deregulation and struc-tural reforms, Abe spent much of 2015 battling to pass unpopular security legislation. Although voters admire Abe for his strong convictions, few share his obsession with security reform, especially when it comes at the expense of the economy. As was the case with Koizumi before him, Abe’s nationalist principles are divisive both at home and abroad. But despite their failings, Koizumi and Abe are the contemporary Japanese prime

ministers to whom all others are compared, and in most cases, found wanting.

6. The LCI in the Japanese Context

When applied to Japanese case studies, two problems arise with the LCI as it is operationalised here. First, bi-ases in the electoral system that benefit larger par-ties—in particular the LDP—complicate the use of poll-ing data on party support relative to the most recent election result as an indicator of a prime minister’s re-lations with the electorate as leader of his party (Table 3, indicator 06). Disparities in population size between electoral districts, which in some cases is as extreme as four to one, benefits the LDP that tends to do well in smaller, rural constituencies (Kabashima & Steel, 2012; Reed, 2003). Electoral pacts and the fact that many smaller parties lack the means to field candidates in every district, leads to substantial tactical voting in SMD ballots. The impact of tactical voting is seen when comparing the vote share received by each party at the 2014 general election in the SMD and PR ballots (Table 6). When casting their PR ballot, voters are more likely to vote for their true preference, as proportional dis-tribution means seats and votes correlate more closely than in the SMD ballot. For this reason, if applying the LCI to Japanese leaders in the future, scholars may pre-fer to compare party polling relative to only the PR election results.

Second, using party support data to gauge the prime minister’s relations with the public is further complicated by the fact that a party and its leader can have very different standing in the public mind. The LDP dominated government for many decades from 1955, with voters returning the party to power whether or not its leader was personally popular. Partly, this was because policy was driven by the rul-ing party and its allies in business and the bureaucra-cy, rather than by the prime minister. Despite a per-sonal approval rating of just 17 per cent, Prime Minister Mori was re-elected in the June 2000 general election, with the LDP receiving 41 per cent of SMD votes and 28 per cent in the PR ballot (Figure 2). Since

Table 6. December 2014 general election results.

SMD PR

% Vote Seats % Vote Seats

Liberal Democratic Party 48.1 222 33.1 68

Democratic Party of Japan 22.5 38 18.3 35

Japan Innovation Party 8.2 11 15.7 30

New Komeito 1.4 9 13.7 26

Japanese Communist Party 13.3 1 11.4 20

Party for Future Generations 1.8 2 2.7 0

Social Democratic Party 0.8 1 2.5 1

People's Life Party 1.0 2 1.9 0

Source: Election Resources (2015). Retrieved from http://www.electionresources.org/jp


Junichiro Koizumi’s dominant leadership changed pub-lic expectations of the prime minister, the fortunes of party and leader have become more intertwined. Pub-lic assessment of the prime minister and his party, however, continue to be somewhat separate. For most of his premiership Koizumi was considerably more popular than the LDP (NHK, 2015). Koizumi used this situation to his advantage, setting himself up in opposi-tion to his own party, a tactic used successfully by Margaret Thatcher and Tony Blair (Hennessy, 2001). Looking at Koizumi’s consistently low rating for party support on the LCI creates a misleading picture of his relationship with the electorate (Table 5). A similar problem is observed for Shinzo Abe, who was also more popular in his second term than his party (Table 3) (NHK, 2015).

For the LCI to better fit the Asian context more broadly, the dynastic nature of the region’s politics must be taken into consideration. Political families are a feature of politics in India, Thailand, Japan and sever-al other Asian states. Among the nine Japanese leaders analysed in this article, four are either the sons or grandsons of former prime ministers. Thought must be given to how to operationalise the benefits conferred by family networks within the LCI. One-party domi-nance is another feature of Japanese politics that is common across Asia. When one party dominates the political scene, LCI indices measuring inter-party sup-port levels may be less important to leadership author-ity than intra-party factional politics.

7. Conclusion

Applying the LCI to Japanese leaders reveals a lack of policy vision and an inability to communicate a clear purpose for seeking power as the underlying causes of short tenure in most cases. Prime ministers lacking the necessary skills to acquire and deploy leadership capital have become more likely to achieve office since changes to the Japanese electoral system in 1994. Electoral reform precipitated the decline of fac-tions within the LDP that had sustained a seniority-based promotion system requiring potential leaders to hone their political skills in a variety of party posi-tions before reaching the premiership. Electoral re-form also eventually brought to power three prime ministers from the DPJ. LDP dominance for much of the post-war period deprived most DPJ politicians of more than fleeting ministerial experience. Outside of government, DPJ leaders were unable to develop the networks within the bureaucracy necessary to facili-tate effective policy-making once in office. In most cases, poor leadership skills quickly translated into falling public popularity, with negative consequences for the prime ministers’ relational and reputational capital. Personal poll ratings of less than 25 per cent were a tipping point from which a leader was unable

to return. Ironically, electoral reforms brought to of-fice leaders with poorly developed skills at the same moment as party leadership became a more im-portant factor in determining voter choice at elec-tions.

Junichiro Koizumi and Shinzo Abe (second term) managed to sustain power for substantially longer than their counterparts by articulating a coherent personal and policy narrative. Koizumi’s leadership capital was enhanced by his charisma and uncommon communica-tion skills. Although Abe worked on his public speaking skills between his first and second governments, he remains a rather underwhelming performer. The lesson here is that the message is more important than the manner in which it is delivered. Koizumi and Abe both also benefited from factors not captured by the LCI. Abe took office after three years of unpopular DPJ rule. Since their election defeat in 2012, the DPJ has been riven with internal disputes (divisions that actually be-gan before the party lost power and were a factor in its defeat), not least over controversial issues such as col-lective self-defence. In the December 2014 general election, the DPJ won just 73 seats, a result only mar-ginally better than when it lost power in 2012. Abe has been able to exploit the weakness of opposition parties to push controversial reforms such as the 2013 Secrecy Act and the reinterpretation of Article 9, policies that were unthinkable in the political context of his first premiership. Koizumi, in contrast, faced a surging op-position, his attacks on his own party helping the DPJ to win new ground. But despite Koizumi’s criticism of his party, both leaders benefited from leading the LDP. Decades of incumbency affords the LDP special lever-age over Japan’s bureaucracy and with smaller political parties. For the latter, coalition with the LDP offers the most assured route to office.

Koizumi’s personal skills were a key factor in sus-taining his leadership capital for so long. His dominant behaviour, however, was less suited to delivering last-ing change. Abe’s reforms have also been partial. Alt-hough he has announced hundreds of structural re-form initiatives, few have been turned into tangible legislation (Kingston, 2014). Despite their difficulties in achieving their agenda, the conviction politics of Koi-zumi and Abe undermine the dominant paradigm that strong leadership is the antitheses of Japanese cultural preferences for consensus and conformity.

Acknowledgments

I am grateful to Sunny Gladish for her research support

for this project.

Conflict of Interests

The author declares no conflict of interests.


References

Anderson, G. (2004). Lionheart or paper tiger? A first-term Koizumi retrospective. Asian Perspective, 28(1), 149-182.

Barber, J. (1972). The presidential character. Englewood Cliffs N.J: Prentice Hall.

BBC. (2000). Profile: Yoshiro Mori. BBC News. Retrieved from http://news.bbc.co.uk/2/hi/asia-pacific/70232 3.stm

Bennister, M., ’t Hart, P., & Worthy, B. (2015). Assessing the authority of political office-holders: The leader-ship capital index. West European Politics, 38(3), 417-440.

Boin, A., & ’t Hart, P. (2003). Public leadership in times of crisis: Mission impossible. Public Administration Re-view, 63(5), 544-553.

Bourdieu, P. (2005). The political field, the social science field and the journalistic field. In R. Benson (Ed.), Bourdieu and the journalistic field (pp. 29-48). Cam-bridge: Polity Press.

Breslauer, G. (2002). Gorbachev and Yeltsin as leaders. Cambridge: Cambridge University Press.

Burke Johnson, R., & Onwuegbuzie A. (2007). Toward a definition of mixed methods research. Journal of Mixed Methods Research, 1(2), 112-133.

Burns, J. M. (1978). Leadership. New York: Haper and Row.

Burns, J. M. (2003). Transformational leadership. Oxford: Oxford University Press.

Burrett, T. (forthcoming). Abe road: Comparing Japanese Prime Minister Shinzo Abe’s leadership of his first and second governments. Parliamentary Affairs. Manuscript submitted for publication.

Bynander, F., & ’t Hart, P. (2006). When power changes hands: The political psychology of leadership succes-sion in democracies. Political Psychology, 27(5), 707-730.

Dahl, R. (1961). Who governs? London: Yale University Press.

Election Resources. (2015). Election resources on the in-ternet: Parliamentary elections in Japan, parts I and II. Election Resources. Retrieved from http://www. electionresources.org/jp

Elgie, R. (1995). Political leadership in liberal democra-cies. London: Palgrave.

Estévez-Abe, E. (2006). Japan’s shift towards a Westmin-ster system. Asian Survey, 46(4), 632-651.

Foley, M. (2009). Gordon Brown and the role of com-pound crisis in the pathology of leadership decline. British Politics, 4(4), 498-513.

George Mulgan, A. (2000). Japan’s political leadership deficit. Australian Journal of Political Science, 3(5), 183-202.

George Mulgan, A. (2003). Japan’s ‘Un-Westminster’ sys-tem: Impediments to reform in a crisis economy. Government and Opposition, 38(1), 73-91.

George Mulgan, A. (2015). Ozawa Ichiro and Japanese politics: Old versus new. Abingdon, UK: Nissan Insti-tute, Routledge Japanese Studies.

Greenstein, F. (2000). The qualities of effective presi-dents: An overview from FDR to Bill Clinton. Presi-dential Studies Quarterly, 30(1), 178-185.

Greenstein, F. (2001). The presidential difference: Lead-ership style from FDR to Clinton. Princeton, NJ: Princeton University Press.

Harding, R., & Lewis, L. (2015, September 9). The third arrow of Abenomics: A scorecard. The Financial Times.

Hayashi, Y. (2011, August 30). Yoshihiko Noda is low-key politician with hawkish bent. Wall Street Journal.

Heffernan, R. (2003). Prime ministerial predominance? Core executive politics in the UK. The British Journal of Politics & International Relations, 5(3), 347-372.

Heifetz, R., Grashow, A., & Linsky, M. (2009). The prac-tice of adaptive leadership. Boston: Harvard Business Press.

Hennessy, P. (2001). The prime minister: The office and its holders since 1945: The job and its holders since 1945. New York: Palgrave.

Herrmann, R. (1988). The empirical challenge of the cognitive revolution: A strategy for drawing infer-ences about perceptions. International Studies Quar-terly, 32(2), 175-203.

Hesse-Biber, S., & Burke Johnson, R. (2013). Coming at things differently: Future directions of possible en-gagement with mixed methods research. Journal of Nixed Methods Research, 7(2), 103-109.

Hesse-Biber, S., & Burke Johnson, R. (2015). The Oxford handbook of multimethod and mixed methods re-search inquiry. Oxford: Oxford University Press.

Hiwatari, N. (2005). Japan in 2004: ‘Courageous’ Koizumi carries on. Asian Survey, 45(1), 41-53.

Inoguchi, T. (2013). Japan in 2012: Voters swing, and swing away soon. Asian Survey, 53(1), 184-197.

Ishibashi, N. (2007). The dispatch of Japan’s self defense forces to Iraq. Asian Survey, 47(5), 766-789.

Kabashima, I., & Steel, G. (2012). Changing politics in Ja-pan. Ithaca, NY: Cornell University Press.

Keller, J. (2005). Leadership style, regime type, and for-eign policy crisis behavior: A contingent monadic peace. International Studies Quarterly, 49(2), 205-231.

King, A. (1991). The British Prime Ministership in the age of the career politician. West European Politics, 14, 25-47.

Kingston, J. (2014, October 18). The Abe conundrum and the pitfalls ahead. The Japan Times.

Krauss, E. (2000). Broadcasting politics in Japan: NHK and television news. Ithaca, NY: Cornell University Press.

Krauss, E., & Nyblade, B. (2005). ‘Presidentialization’ in Japan? The prime minister, media and elections in Japan. British Journal of Political Science, 35(2), 357-368.


Krauss, E., & Pekkanen, R. J. (2004). Explaining party ad-aptation to electoral reform: The discreet charm of the LDP? Journal of Japanese Studies, 30(1), 1-34.

Krauss, E., & Pekkanen, R. J. (2011). The rise and fall of Japan’s LDP: Political party organizations as historical institutions. Ithaca, NY: Cornell University Press.

Laing, M., & ’t Hart, P. (2011). Seeking and keeping the hot seat. In P. ‘t Hart & J. Uhr (Eds.), How power changes hands (pp. 111-132). Basingstoke, UK: Pal-grave.

Lasswell, H. (1936). Politics: Who gets what, when and how. New York: Peter Smith.

Machidori, S. (2006, October). ‘Tsuyoi Shusho’ ha Nichijo to Naru’ [A ‘strong prime minister’ will be usual]. Chuo Koron, pp. 174-184.

Matsumoto, M. (2001). Seiji Ishiki Zusetsu [Illustrative explanation of political consciousness]. Tokyo, Japan: Chuo Koron.

McCall Rosenbluth, F., & Thies, M. (2010). Political change and economic restructuring. Oxford: Prince-ton University Press.

Mishima, K. (2007). Grading Japanese prime minister Koizumi’s revolution. Asian Survey, 47(5), 727-748.

Mishima, K. (2012). Explaining Japan’s fragile premier-ships in the post-Koizumi era. East Asia, 29, 275-293.

National Diet of Japan. (2012). The Fukushima nuclear accident independent investigation commission. To-kyo, Japan: The National Diet of Japan.

Neustadt, R. (1991). Presidential power and the modern presidents. Oxford: Macmillan.

NHK. (2015). Seiji Ishiki Getsurei Chōsa [Political aware-ness monthly survey]. NHK Online. Retrieved from http://www.nhk.or.jp/bunken/yoron/political/index.html

Nye, J. (2008). Powers to lead: Soft, hard, and smart. Ox-ford: Oxford University Press.

Otake, H. (2006). Koizumi Junichiro popyurizumu no kenkyu [Study of Junichiro Koizumi’s populism]. To-kyo, Japan: Toyo Keizai Shinpo.

Pekkanen, R., & Krauss, E. (2005). Japan’s ‘coalition of the willing’ on security policies. Orbis, 49(3), 429-444.

Rebuild Japan Initiative Foundation. (2013). DPJ admin-istration: Challenges and failures. Tokyo, Japan: Chu-okoron-Shinsha.

Reed, S. (2003). Japanese electoral politics: Creating a new party system. London, UK: Routledge.

Schmidt, C. (2011). The DPJ and its factions: Benefit or threat? Hitotsubashi Journal of Social Studies, 43, 1-21.

Schoppa, L. J. (2011). The evolution of Japan’s party sys-tem: Politics and policy in an era of institutional change. Toronto, Canada: University of Toronto Press.

Shimizu, M. (2005). Kantei shudō—Koizumi Junichirō no kakumei [Junichiro Koizumi’s palace revolution]. To-kyo, Japan: Nihon Keizai Shimbun.

Shinoda, T. (2005). Japan's cabinet secretariat and its emergence as core executive. Asian Survey, 45(5), 800-821.

Shinoda, T. (2011). Prime ministerial leadership. In A. Gaunder (Ed.), Routleadge handbook of Japanese politics (pp. 48-59). Abingdon, UK: Routledge.

Skowronek, S. (1993). The politics president’s make. New Haven, CT: Yale University Press.

Stockwin, J. A. A. (2008). Governing Japan: divided poli-tics in a resurgent economy. Oxford: Wiley-Blackwell.

Takayasu, K. (2010). Naze Nihon No Shusho Ha Suguni Kotaisurunoka [Why do the Japanese prime ministers change often?]. Sekai, 801, 103-112.

Takenaka, H. (2006). Shusho shihai: nihon seiji no henbo [Prime ministerial dominance: The transformation of Japanese politics]. Tokyo, Japan: Chuo Koron Shinsha.

The Economist. (2006a, August 17). Drawing Yasukuni’s sting. The Economist.

The Economist. (2006b, September 14). The man who remade Japan. The Economist.

The Economist. (2011a, April 16). Flailing: Japan’s post-tsunami politics. The Economist.

The Economist. (2011b, September 3). The mud-lover: Japan’s new leader. The Economist.

The Economist. (2012, July 28). Eyes right: Politics in Ja-pan. The Economist.

Thompson, J. B. (2000). Political scandal. Cambridge: Cambridge University Press.

Tucker, R. (1981). Politics of leadership. Columbia: Uni-versity of Missouri.

Uchiyama, Y. (2010). Koizumi and Japanese politics: Re-form strategies and leadership style. London: Routledge.

Yoshida, R. (2014, July 11). Wording of questions may color attempts to gauge public support. Japan Times.

About the Author

Tina Burrett is Associate Professor of Political Science at the Faculty of Liberal Arts, Sophia University, Japan. She holds a PhD in Political and Social Sciences from Cambridge University. Her recent publica-tions include Television and Presidential Power in Putin’s Russia (Routledge, 2013), “Russia’s Compet-ing Nationalisms and Relations with Asia” in Asian Nationalisms Reconsidered, Jeff Kingston (Ed.) (Routledge, 2015) and Prime Ministerial Leadership in Britain and Japan (Palgrave Macmillan, 2017, forthcoming).


Appendix

Table A1. Route to power and longevity in office.

Prime Minister Days in Office Route to Power

Mori 386 Internal Party Election Koizumi 1979 Internal Party Election Abe 1 365 Internal Party Election Fukuda 364 Internal Party Election Aso 357 Internal Party Election Hatoyama 256 General Election Kan 451 Internal Party Election Noda 481 Internal Party Election Abe 2 1070* General Election

Note: * Until 31 December 2015.

explaining japan’s revolving door premiership: applying

Documents